Many animal and plant diseases, such as the human diseases cholera, typhoid fever, and legionelliosis, are caused by facultative bacterial pathogens, that exist and proliferate outside the host environment. While the epidemiology of obligate pathogens is well described by established population models of disease, these models are inadequate for facultative bacterial pathogens, and do not incorporate several important features of their biology. The frequency of virulent strains, for example, is strongly affected by local scale competition, both within the host, where virulent strains have a competitive advantage, and outside the host, where the cost of virulence is expected to be disadvantageous. Moreover, virulence genes are often encoded on mobile elements such as plasmids, allowing their spread among bacteria through horizontal gene transfer. In this proposal, modeling and empirical approaches are combined to investigate the influence of local scale competition and horizontal gene transfer on virulence dynamics, and the maintenance of virulence, for facultative bacterial pathogens. General models will be developed, parameterized and tested using Agrobacterium tumefaciens, an easily manipulated and ubiquitous facultative plant pathogen. A. tumefaciens causes crown gall disease, and is a model for pathogenesis mechanisms and horizontal gene transfer. Construction of isogenic lines with defined mutations, will allow unambiguous evaluation of the costs and competitive advantages conferred by virulence and conjugation. Moreover, several defining aspects of the ^. tumefaciens disease environment, such as plant-released inducers of virulence and production of customized nutrients, called opines, by infected plants, can be readily simulated in the laboratory. Conditions that should favor virulent strains or avirulent strains (nonpathogens and cheaters), will be tested for the predicted interactions. The effects of resource competition will be evaluated in the diseased host and non-host environments, performing experiments in vitro, in vivo, and the field. These results then form the basis for evaluating the importance of horizontal gene transfer of virulence genes and for scaling up to infection dynamics and disease incidence.